WebSocket

A single connection for multiple concurrent text-to-speech generations, with input and output streaming. Provides the best latency and performance out of the text-to-speech endpoints.

HandshakeTry it

WSS
wss://api.respeecher.com/v1/public/tts/en-rt/tts/websocket

Send

generateobjectRequired

Text-to-speech generation request.

OR
cancelobjectRequired
Stop receiving audio for a particular context ID. Credits for already generated yet unstreamed audio will still be charged.

Receive

responseobjectRequired
Audio chunk, finished context notification, or an error.